Regime-Aware Interpretable Ranking for Sector Rotation

By Kelvin J. L. Koa, Ziling Liu, Xinyang Li, Shuowei Feng, Ruoyu Liang, Mathieu Giroux, Eugene Y. Wen, Xuan Yao, Ke-Wei Huang, Hugo Lamarre

Rating

1851
Battle Count: 50

Relevance

9/10
Highly relevant for institutional quantitative strategies focusing on sector rotation and benchmark-relative active management. The focus on interpretability and regime-awareness addresses key practical constraints for portfolio managers.

Implementation Complexity

8/10
High complexity due to the multi-stage architecture involving temporal projections, cross-attention for regime conditioning, self-attention for sector dependencies, and specific listwise ranking objectives. Requires careful handling of financial data leakage via purged cross-validation.

Reproducibility

3/5
The paper provides detailed architectural specifications, hyperparameters, and data sources (Bloomberg). However, no public code repository URL is provided in the text, and Bloomberg data is proprietary.

Abstract

Sector rotation is a common decision in benchmark-relative portfolio management, where portfolio managers rank sectors for overweight and underweight positions. We formulate this task as ranking sector actions by expected future risk-adjusted active performance, measured with forward sector-benchmark Sharpe ratios. Such rankings should preserve temporal information, express model outputs through economically familiar signal characteristics, account for changes in signal relevance across macroeconomic environments, and capture dependencies among sectors. We propose RegimeRank, an interpretable sector-ranking model that represents fundamental and technical signals through expert-defined temporal characteristics, conditions them on learned macro factors via cross-attention, and models cross-sector dependencies through self-attention. Expected-gradient attributions identify the contributions of individual signal-transformation pairs, while macroeconomic and cross-sector attention maps provide diagnostics of the learned conditioning and relational structure. We evaluate RegimeRank on weekly S&P 500 data from 2002 to 2024 under combinatorial purged cross-validation, where it outperforms representative ranking and machine-learning baselines across multiple ranking metrics. Ablation studies support the contributions of latent macro states, regime conditioning, and cross-sector modelling, while attribution and attention analyses illustrate the temporal characteristics, macro states, and sector relationships underlying its rankings.

Links

arXiv:2610.26456 ยท Paper rankings

The interactive Everscope explorer (charts, battles, favorites) loads below.